Optimization and Modeling of Flow Characteristics of Low-Oil DDGS Using Regression Techniques

Abstract

Storage conditions, such as temperature, relative humidity (RH), consolidation pressure (CP), and time, affect the flow behavior of bulk solids such as distillers dried grains with solubles (DDGS), which is widely used as animal feed by the U.S. cattle and swine industries. The typical dry-grind DDGS production process in most corn ethanol plants has been adapted to facilitate oil extraction from DDGS for increased profits, resulting in production of low-oil DDGS. Many studies have shown that caking, and thus flow, of regular DDGS is an issue during handling and transportation. This study measured the dynamic flow properties of low-oil DDGS. Flow properties such as stability index (SI), basic flow energy (BFE), flow rate index (FRI), cohesion, Jenike flow index, and wall friction angle were measured at varying temperature (20°C, 40°C, 60°C), RH (40%, 60%, 80%), moisture content (MC; 8%, 10%, 12% w.b.), CP (generated by 0, 10, and 20 kg overbearing loads), and consolidation time (CT; 2, 4, 6, 8 days) for low-oil DDGS. Response surface modeling (RSM) and multivariate analysis showed that MC, temperature, and RH were the most influential variables on flow properties. The dynamic flow properties as influenced by environmental conditions were modeled using the RSM technique. Partial least squares regression yielded models with R2 values greater than 0.80 for SI, BFE, and cohesion as a function of MC, temperature, RH, CP, and CT using two principal components. These results provide critical information for quantifying and predicting the flow behavior of low-oil DDGS during commercial handling and transportation.

Description

Citation: R. Bhadra, R. P. K. Ambrose, M. E. Casada, S. Simsek, K. Siliveru. (2017). Optimization and Modeling of Flow Characteristics of Low-Oil DDGS Using Regression Techniques. Transactions of the ASABE. 60(1): 249-258. (doi: 10.13031/trans.11928)

Keywords

Dynamic flow properties, Flowabilty, Low-oil DDGS, Multivariate modeling

Citation